The trial most often quoted for azelaic acid and melasma gave every one of its 329 participants a broad-spectrum sunscreen — in both groups — and included no group that used sunscreen alone. It is a real, large, double-blind study, and it still cannot tell you how much of what happened belonged to the ingredient.
That gap between there is a study and the study answers the question is where most ingredient marketing lives. Azelaic acid is a useful molecule to learn it on, because it has both: genuinely strong evidence in one condition, and much thinner evidence in the one people usually ask about. See Cellular Clarity Serum and the Cellular Clarity science page.
What is azelaic acid?
Azelaic acid is a non-phenolic saturated dicarboxylic acid with nine carbon atoms, produced naturally on human skin by the yeast Malassezia. It is a licensed medicine as well as a cosmetic ingredient — the US Food and Drug Administration has approved it for papulopustular rosacea in a 15% gel formulation.
A 2024 narrative review in Clinical, Cosmetic and Investigational Dermatology collects the activities attributed to it: antibacterial, anti-keratinising, antioxidant, anti-inflammatory and pigment-related (PubMed 39464747). Being a narrative review, it summarises existing work rather than testing anything new, and a good deal of the mechanistic detail it gathers comes from laboratory work rather than from people.
What is the strongest evidence for azelaic acid?
Rosacea, by a wide margin. The Cochrane review Interventions for rosacea (Cochrane Database of Systematic Reviews, 2015) pooled 106 studies and 13,631 participants and found azelaic acid more effective than placebo in papulopustular rosacea, rating that body of evidence high quality — drawn from four trials in which the participants themselves assessed the change (PubMed 25919144).
The limits belong in the same breath. That review was restricted to moderate-to-severe rosacea, and of its 106 included studies only 12 were judged at low risk of bias — 57 were unclear and 37 were high. A separate phase 3 trial of azelaic acid foam at 15% against its own vehicle beat the vehicle on both co-primary endpoints at P < 0.001 (Cutis, 2015; PubMed 26244354), but that is a manufacturer's formulation trial of a prescription-strength product in people with a diagnosis.
What did the melasma trials actually find?
They found that 20% azelaic acid was not measurably different from 4% hydroquinone — which is a comparison between two active treatments, not a demonstration that either beat nothing. In the largest of them, 329 women used one cream or the other for 24 weeks under double-blind conditions; azelaic acid produced good or excellent results in 65% of them, with no significant difference between the two treatments on overall rating, lesion size or pigmentary intensity (International Journal of Dermatology, 1991; PubMed 1816137).
Here is the design detail that rarely survives into a product page: both groups were also given a broad-spectrum sunscreen, and there was no arm that used sunscreen alone. Since sun exposure is the main driver of melasma, a trial built this way cannot separate what the cream did from what the sunscreen did. The study is not bad — it reports what it found honestly, including that azelaic acid caused none of the severe reactions hydroquinone can. It simply answers a narrower question than the one it gets quoted for.
| Study | Strength tested | Design | What it found | What it cannot tell you |
|---|---|---|---|---|
| Cochrane review, 2015 | Prescription strengths | 106 trials pooled, 13,631 people | Better than placebo in papulopustular rosacea; high-quality evidence | Anything about mild rosacea, or about cosmetic strengths |
| Phase 3 foam trial, 2015 | 15% foam | Randomised, double-blind, vehicle-controlled | Beat vehicle on both co-primary endpoints | Whether it would beat an active competitor; industry-run |
| Melasma trial, 1991 | 20% cream | Double-blind, 329 women, 24 weeks, active comparator | No significant difference from 4% hydroquinone | How much came from the sunscreen both groups used |
| Adult acne maintenance, 2023 | Cosmetic strength | Randomised, investigator-blinded, 53 women, 16 weeks | Fewer relapses, but no statistical difference in severity score | Whether the azelaic acid or the other ingredient mattered |
Does cosmetic-strength azelaic acid do the same thing?
On the evidence available, that has not been shown — and the one trial designed to test it is worth reading closely. A 2023 study in the Journal of the European Academy of Dermatology and Venereology gave 53 adult women either a cosmetic containing Myrtus communis extract and azelaic acid, or a plain light moisturising cream, twice daily for 16 weeks, after their acne had already been cleared by treatment (PubMed 36729399).
Relapses were fewer in the cosmetic group — three subjects against eight. But on the investigator's global assessment, the study's main measure of severity, there was no statistical difference between the cosmetic and the plain moisturiser. The authors describe the study as exploratory, it was small, and because the test product contained two actives it cannot tell you which one, if either, was responsible. This is the honest counterpoint to everything above: prescription-strength evidence does not travel down to cosmetic strength on its own.
What about the derivative, potassium azeloyl diglycinate?
It is a different molecule from azelaic acid, and it has to earn its own evidence. Potassium azeloyl diglycinate — sold as Azeloglicina and used in cosmetics for its water solubility — is a derivative, and a trial of the parent compound is not a trial of it.
Its best human study is a randomised, double-blind, placebo-controlled pilot: 42 people with stage I–II rosacea, 28 on the cream and 14 on placebo, twice daily for four weeks. Erythema fell on both instrument readings and clinical assessment (Journal of Cosmetic Dermatology, 2012; PubMed 22360333). Note the constraints: it is a pilot, it is small and short, and the cream paired 5% potassium azeloyl diglycinate with 1% hydroxypropyl chitosan — so the derivative was never tested by itself. Beyond that, much of the published work on it is in vitro, meaning cells in a dish rather than people (Giornale Italiano di Dermatologia e Venereologia, 2017; PubMed 27013148).
How do you judge an ingredient's evidence in four questions?
- What was it compared with? Against nothing, against a vehicle, or against another active treatment. Each answers a different question, and only the first two tell you the ingredient did something.
- What else was everyone given? If both groups also used a sunscreen, a moisturiser or a cleanser, the trial measures the ingredient plus that — not the ingredient.
- At what strength? Evidence attaches to a concentration. A result at 15% or 20% says nothing about 2%.
- Is it the same molecule? A derivative is a new compound. It starts its evidence again from zero.
If you want to go further on that last habit, we wrote a longer piece on what “clinically studied” actually means on a label, and one on how vitamin C serums are studied. For the practical end of things, there is our guide to layering serums in an AM and PM routine.
What we make
We make a serum, Cellular Clarity Serum, which contains potassium azeloyl diglycinate (Azeloglicina) at 2% alongside Delentigo at 4%. None of the research described in this article was carried out on that serum: the rosacea and melasma evidence concerns azelaic acid at prescription strengths of 15% and 20%, and the derivative's pilot study used 5% in combination with another ingredient. It contains soy-derived isoflavones — if you are sensitive to soy, this is the line that matters to you.
Frequently asked questions
Is azelaic acid an exfoliating acid like glycolic or salicylic acid?
No. It belongs to a different chemical family — a nine-carbon dicarboxylic acid — and the 2024 review above lists antibacterial, anti-keratinising, antioxidant, anti-inflammatory and pigment-related activities for it rather than exfoliation. Sharing the word “acid” does not make two ingredients behave alike.
Why is prescription azelaic acid 15% or 20% when cosmetics use much less?
Because those are the strengths the trials used, and a medicine has to be licensed at the strength it was tested at. Cosmetic regulation works differently, which is why a cosmetic can contain the same ingredient without carrying the same evidence.
Does azelaic acid work for melasma?
The 1991 trial found 20% azelaic acid no different from 4% hydroquinone over 24 weeks, with both groups also using sunscreen. That is a genuine result against an active comparator, but the absence of a sunscreen-only group means the trial cannot isolate the cream's own contribution.
Is potassium azeloyl diglycinate the same as azelaic acid?
No — it is a derivative, a different molecule made to be water-soluble. Its human evidence is one small four-week pilot in rosacea, in which it was combined with another ingredient.
Where should I start if I am reading a product page?
Look for the concentration next to the ingredient name. If a page names an active but not its percentage, you cannot check it against any study, and neither can anyone else.
Written by the VyVa Editorial Team. Every study named here was checked against its PubMed record before publication, and its limits are stated alongside its finding.
This article is about cosmetic ingredients and published research. VyVa products are cosmetics intended to improve the appearance of skin; they are not intended to diagnose, treat, cure or prevent any disease, and nothing here should be read as medical advice. Rosacea, melasma and acne are medical conditions — speak to a doctor or dermatologist about them. Patch-test any new product on the inner arm for 48 hours before use, and stop if irritation occurs. Cellular Clarity Serum contains soy-derived isoflavones.
